Search: “safety”

arXiv:2609.27900v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly deployed in multi-agent systems where a principal agent decomposes tasks and delegates them to subordinate

arXiv:2609.22818v1 Announce Type: cross Abstract: Memory-poisoning defenses for LLM agents are typically evaluated by their ability to prevent attacks. However, the traffic they process is rarely adve

arXiv:2609.19630v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly integrated into vehicle voice assistants. But linking natural-language requests to vehicle functions creat

arXiv:2609.17292v1 Announce Type: new Abstract: Control barrier functions for input-constrained systems place the admissible input set inside the definition of the safe set, yet the resulting barrier

Interesting safety paper from Microsoft. They find that a weaker, unaligned model can split a harmful task into harmless-looking subquestions, ask an aligned frontier model each one in a separate sess

arXiv:2512.01037v3 Announce Type: replace-cross Abstract: As safety alignment becomes standard in large language models, refusal behavior has become an important part of model reliability. However, mo

arXiv:2609.13866v1 Announce Type: new Abstract: Group-relative policy optimization (GRPO and descendants) can discard no-contrast rollout groups through dynamic sampling, while practical implementatio

arXiv:2609.09793v1 Announce Type: cross Abstract: Directional ablation removes an aligned language model's ability to refuse by projecting a single 'refusal direction' out of the weights that write th

arXiv:2608.30201v1 Announce Type: new Abstract: Directly robustifying small-signal stability in AC optimal power flow is challenging since the stability boundary in the original uncertainty space is i

arXiv:2608.21570v1 Announce Type: new Abstract: Deploying a safety layer for large language models on commodity hardware is constrained by the guards available to do it: current open guard models hold

arXiv:2608.22335v1 Announce Type: cross Abstract: Bengali is the seventh-most-spoken language globally, yet LLM safety evaluation remains overwhelmingly English-centric. We introduce BanglaSafe, a ben

arXiv:2608.19579v1 Announce Type: new Abstract: Large Language Models (LLMs) are increasingly deployed in high-stakes applications, yet their tendency to generate toxic, harmful, or policy-violating c

arXiv:2608.16041v1 Announce Type: new Abstract: We introduce ScenarioCharacterization, an open-source framework for automated, dataset-agnostic profiling of driving scenarios in trajectory datasets. O

arXiv:2608.09443v1 Announce Type: new Abstract: Large language model (LLM) agents can support medication review between clinical visits, but safe choices for older adults with multimorbidity depend on

arXiv:2508.03875v2 Announce Type: replace Abstract: Many sequential decision problems offer qualitatively different ways of influencing the environment: some interventions act immediately, whereas oth

arXiv:2601.17003v2 Announce Type: replace-cross Abstract: Mental-health AI safety is typically evaluated with small, simulation-based benchmarks that may not reflect the linguistic and contextual dive

arXiv:2509.23690v2 Announce Type: replace-cross Abstract: Safety hazards in the home are a leading cause of preventable domestic injuries, motivating an automated inspector that actively explores a ho

arXiv:2608.03201v1 Announce Type: new Abstract: Safety guards are widely used to filter harmful content and are typically trained via supervised fine-tuning on labeled prompt-response pairs. We audit

Mistral AI introduced Shieldstral, a 3‑billion‑parameter, open‑weights model designed for content‑safety tasks and capable of on‑device deployment. The announcement was shared via a tweet from @Mistra

When the phrase 'OpenAI hacked Hugging Face' has more or less entered mainstream culture, you know we have an AI problem. This week, we learned more about exactly how OpenAI's agent broke out of a san

arXiv:2403.18591v3 Announce Type: replace-cross Abstract: Broadcast protocols are programs designed to be executed by networks of processes. Each process runs the same protocol, and communication betw

arXiv:2607.26518v1 Announce Type: new Abstract: Reliable visual safety understanding in real-world scenarios demands more than just object recognition; it requires causal reasoning under epistemic unc

Earlier this month, OpenAI gave several of its AI models a task: complete a test designed to measure their cybersecurity capabilities. It put the systems in a sandboxed environment without an internet

arXiv:2607.22929v1 Announce Type: new Abstract: A short fine-tuning run can undo the safety guards of an open-weight model---retraining a refusal-trained assistant to aid weapons development or produc

arXiv:2510.24411v3 Announce Type: replace Abstract: Computer-using agents powered by Vision-Language Models (VLMs) have demonstrated human-like capabilities in operating digital environments like mobi

arXiv:2607.19366v1 Announce Type: new Abstract: Safety as Polytope (SaP) learns linear half-space constraints in LLM hidden space but requires per-category tuning of the constraint count K. We show th

arXiv:2607.20286v1 Announce Type: cross Abstract: We propose a novel framework for computing rigorous bounds on the probability that a large language model (LLM) generates harmful output to a given pr

arXiv:2607.12469v1 Announce Type: cross Abstract: Many agent-safety evaluation results are not yet load-bearing evidence: identical nominal outcomes (task success, attack success, monitor scores) may

arXiv:2607.04234v1 Announce Type: cross Abstract: Deformable object manipulation poses challenges beyond task completion: successful execution must also maintain safe physical interaction, holding the

arXiv:2512.11891v2 Announce Type: replace Abstract: Vision-Language-Action (VLA) models have demonstrated remarkable capabilities in generalizing across diverse robotic manipulation tasks. However, de

arXiv:2602.08653v2 Announce Type: replace Abstract: Quadrotor unmanned aerial vehicles (UAVs) are increasingly deployed in complex missions that demand reliable autonomous navigation and robust obstac

arXiv:2606.30049v1 Announce Type: new Abstract: Visibility distance is critical to maritime navigational safety because it determines the effective observation range of shipborne and shore-based monit

arXiv:2606.26406v1 Announce Type: cross Abstract: We propose a complete architectural blueprint for safe artificial general intelligence based on a closed reentry loop (D I cycle). In contrast to feed

arXiv:2606.20651v1 Announce Type: cross Abstract: Autonomous drone swarms in space-constrained environments such as warehouses, inspection corridors, and urban delivery routes must share limited airsp

A car pulls up to the curb. The app says, “Your ride is here.” No one’s in the driver’s seat. For people who live in one of the dozens of cities now hosting robotaxi services, this is already a realit

This article discusses Claude Fable 5, likely exploring Anthropic's latest version of their AI model and examining new fables or narratives related to AI safety concepts. The piece probably analyzes h

arXiv:2606.07929v1 Announce Type: new Abstract: Large language models (LLMs) are entering clinical practice based on benchmark accuracy that may fail to detect safety-relevant failure modes. Here we p

arXiv:2605.25806v2 Announce Type: replace Abstract: Women's safety and security are paramount for a modern society. Crimes against women occur in daylight as well as in low-light conditions. Often, su

arXiv:2606.00813v1 Announce Type: cross Abstract: Safety alignment in LLMs does not improve monotonically across model generations. Studying four generations of Google's Gemma family (7B-31B) with qua

arXiv:2606.02444v1 Announce Type: new Abstract: Recent evidence shows that people with eating disorders (EDs) are increasingly seeking guidance, advice, and emotional support from Large Language Model

arXiv:2512.10234v2 Announce Type: replace-cross Abstract: Even LLMs that appear safe during evaluation can still produce harmful responses in deployment. Because stochastic sampling yields different r

arXiv:2606.01970v1 Announce Type: new Abstract: Reliable autonomous UAV swarms in Search and Rescue (SAR) missions require fault-tolerant coordination capable of sustaining operations despite agent de

arXiv:2605.29224v1 Announce Type: cross Abstract: AI agents augment large language models with external tools such as web retrieval, enabling grounded and up-to-date responses. However, incorporating

arXiv:2603.08761v2 Announce Type: replace-cross Abstract: We argue that formal certification of AI alignment over open-ended or unbounded input domains is impossible under standard assumptions in comp

arXiv:2605.22373v1 Announce Type: cross Abstract: Safety classifiers are essential safeguards within generative AI systems, filtering harmful content or identifying at-risk users when interacting with

arXiv:2605.16894v1 Announce Type: new Abstract: Reinforcement Learning (RL) uses rewards to guide learning, yet reward design is typically hand-crafted using heuristics that can be difficult to tune.